The Molecular Therapeutics Section conducts both clinical and laboratory studies in drug resistance. Clinical studies center primarily around reversal of P-glycoprotein-mediated resistance, while the laboratory focus of the section this past year has been on increasing our understanding of non-P-glycoprotein mechanisms of drug resistance. We have examined a novel mechanism of drug efflux and resistance in human breast and colon cancer cells; the mechanism of action of depsipeptide FR901228, and have investigated the intrinsic mechanisms of resistance in renal cell carcinomas. We have studied a variety of compounds in an effort to target these differing mechanisms of resistance. Laboratory studies have identified overexpression of a new non-P-glycoprotein ATP-dependent transporter, termed MXR1 for mitoxantrone resistance gene in a doxorubicin-resistant breast cancer subline and a mitoxantrone-resistant colon cancer subline. In collaboration with the laboratory of Dr. Tito Fojo, this gene was cloned and characterized. MXR has particular activity for mitoxantrone, daunorubicin, doxorubicin, and CPT 11. The gene is localized to chromosome 4, is amplified in the breast cancer cells and rearranged in the colon cancer cells. The gene encodes a predicted half transporter molecule; a type of transporter never previously linked with drug resistance. Using a polyclonal antibody raised against a peptide fragment predicted from the sequence, immunoblot analysis has demonstrated overexpression of a 72 kDa protein in the two sublines. Localization studies are underway in order to determine whether this is a plasma membrane transporter, or whether an intracellular compartment is involved. Studies this year demonstrated the capacity for glucuronidation in these resistant cells; MXR may be involved in metabolism of drug instead of a simple membrane efflux mechanism as for P-glycoprotein. Efflux cannot be reversed by standard P-glycoprotein antagonists; however, two agents able to inhibit MXR-mediated resistance are under study with an aim to develop them for clinical trials. Our focus on intrinsic resistance in renal cell carcinoma has increased. We have completed evaluation of a series of compounds with renal selectivity identified through the NCI drug screen. These compounds are active in cell lines derived from both primary and metastatic renal cell carcinoma, and do not depend upon growth for cytotoxicity. One of these compounds has shown efficacy in the in vivo xenograft model established by the NCI drug screen. Studies in the coming year will emphasize developing this type of agent for clinical trials. We have expanded studies with the depsipeptide FR901228. We are currently studying this compound in a Phase I trial, and concurrently in the laboratory. The compound is a histone deacetylase inhibitor and induces a mitotic arrest in susceptible cells. Both a G1 and a G2 arrest are identified; we have determined that the G1 arrest is p21-dependent. The G2 arrest occurs in prometaphase following chromosomal condensation, and chromosomes do not attach properly to the mitotic spindle. Currently, the molecular basis of these abnormalities is being sought. The clinical focus for the Molecular Therapeutics Section continues to be on Phase I clinical trials. A Phase I trial designed to determine the maximum tolerated dose (MTD) for the cyclosporine analogue, PSC 833, in combination with paclitaxel is complete. A Phase I trial of infusional PSC 833 in combination with infusional vinblastine in renal cell cancer was modified in an effort to increase the vinblastine dose. That trial is spurred by responses noted in a previous Phase I trial. We have conducted a Phase I study of depsipeptide, and identified the dose limiting toxicities as fatigue and thrombocytopenia. EKG changes and a cardiac arrythmia have also been identified. Depsipeptide is transported very efficiently by Pgp, making it likely that the drug will be ineffective in patients whose tumors express Pgp. Thus, studies are planned combining a Pgp antagonist with depsipeptide. In addition, we have developed an assay to determine the effect of depsipeptide on tumor tissue, measuring the change in histone acetylation in turmor and in peripheral T cells, following treatment with depsipeptide. Finally, our protocol combining the ErbB2 inhibitor, Herceptin, with paclitaxel has been opened. The goal of this study is to evaluate changes in molecular signaling following treatment with Herceptin. Laboratory support for the clinical trials includes analysis of biopsies from patients on study for MDR-1 and MRP expression. An assay has been developed using normal T cells from patients receiving PSC 833, which demonstrates the degree of rhodamine efflux inhibition by PSC 833 as a surrogate marker for Pgp antagonism. These studies have tracked the efficacy of Pgp antagonism as we have modified the dose of PSC 833 in the vinblastine interaction trial. As noted above, we have also examined the mechanism of action of depsipeptide in laboratory models. These studies have led to the development of an ex vivo assay for Depsipeptide in which cell cycle arrest and histone acetylation is being examined in patient samples from the Phase I trial. In summary, the focus of the Molecular Therapeutics Section has been the identification of mechanisms of drug resistance, and approaches to overcome them. As long as chemotherapy remains in the anticancer armamentarium, drug resistance will be a problem. Studies have shown that many of the agents developed for new molecular targets are susceptible to the same drug efflux mechanisms identified for the old agents. For example, the newly approved topoisomerase I inhibitors, Topotecan and CPT 11 are substrates for MXR, the transporter encoded by the new mitoxantrone resistance gene. Thus, resistance modulation continues to be a critical area for cancer treatment.